Testing of antiviral agents for the treatment of ocular herpes infections will continue. Drug combinations will be studied. The drugs will be tested in vitro, and the most promising combinations, i.e., those that show synergistic effects against various strains of herpesvirus in tissue culture, will then be evaluated against herpetic disease in the eyes of New Zealand white rabbits. The rabbit model of epithelial herpetic keratitis induced by means of the McKrae strain herpesvirus will be used. Drugs to be evaluated for efficacy and safety will include acyclovir, dihydroxy propyl guanosine, FMAU, and FIAC. Toxicity will be explored in terms of topical, oral, and systemic administration. Light and electron microscopy will be employed to assess cellular toxicity. Drug combinations that show synergistic activity will be tested for efficacy against herpetic stromal disease and iritis in the rabbit eye, as well as the ability to prevent virus shedding in tears, recurrent disease, and the establishment of latency. A new model for herpetic retinitis has been developed, and some of the drug combinations will be tested against this condition. Recombinant interferons, including human leukocyte subspecies, fibroblast interferon, and immune interferons, will also be evaluated in rabbits and owl monkeys, alone and in combination with various antiviral drugs. Non-steroidal anti-inflammatory drugs, such as fluboprofen and suprofen, and immunosuppressants, such as cyclosporine, will be tested for use in the treatment of stromal disease and iritis, to determine if a replacement can be found for corticosteroids that will be less toxic and have fewer side effects. Mechanisms of virulence and neurotropism of the herpesvirus strain will be investigated, including examination of strains that cannot be reactivated from the latent state, as possible immunizing strains, and the study of mutant strains with varying levels of thymidine kinase production, in order to determine if the thymidine kinase gene could be the primary determinant of virulence. Finally, the relationship between substance P-containing ganglionic neurons and neurons harboring herpesvirus in the latent state will be examined, with the long-term goal of identifying neuronal triggers of virus reactivation.